Research map / Graph learning / Graph self-supervised learning
SFGA: Similarity-Constrained Fusion Learning for Unsupervised Anomaly Detection in Multiplex Graphs
Huiliang Zhai, Xiangyi Teng, Jing Liu
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add inter-relation attention (ATT) (ablation) | SFGA · YelpChi | AUROC: 59.63 -> 70.06 | Improved |
| add inter-relation attention (ATT) (ablation) | SFGA · YelpChi | AUPRC: 19.83 -> 24.68 | Improved |
| add similarity constraint (SC) (ablation) | SFGA · YelpChi | AUROC: 67.99 -> 70.06 | Improved |
| add similarity constraint (SC) (ablation) | SFGA · Amazon | AUPRC: 13.17 -> 16.39 | Improved |
| add hierarchical anomaly detector (HD) (ablation) | SFGA · YelpChi | AUROC: 67.82 -> 70.06 | Improved |
| add hierarchical anomaly detector (HD) (ablation) | SFGA · Amazon | AUPRC: 15.79 -> 16.39 | No clear effect |
| combine ATT, SC and HD (ablation) | SFGA · YelpChi | AUROC: 51.44 -> 70.06 | Improved |
| vary alpha (loss weight) from 0 to 1 (ablation) | SFGA · Amazon | AUROC/AUPRC | Mixed |
| vary beta (score weight) from 0 to 1 (ablation) | SFGA · Amazon | AUROC/AUPRC | Mixed |
| multiplex SFGA vs homogeneous GAD baselines | SFGA · Freebase | AUROC: 84.89 -> 92.98 | Improved |
| multiplex SFGA vs TAM | SFGA · Amazon | AUPRC: 34.19 -> 16.39 | Hurt |
Closest papers by meaning
- ADA-GAD: Anomaly-Denoised Autoencoders for Graph Anomaly Detection (AAAI 2024 · similarity 0.89)
- UniFORM: Towards Unified Framework for Anomaly Detection on Graphs (AAAI 2025 · similarity 0.89)
- UniGAD: Unifying Multi-level Graph Anomaly Detection (NeurIPS 2024 · similarity 0.88)
- DiffGAD: A Diffusion-based Unsupervised Graph Anomaly Detector (ICLR 2025 · similarity 0.88)
- Unifying Unsupervised Graph-Level Anomaly Detection and Out-of-Distribution Detection: A Benchmark (ICLR 2025 · similarity 0.87)
- Leveraging Diffusion Model as Pseudo-Anomalous Graph Generator for Graph-Level Anomaly Detection (ICML 2025 · similarity 0.87)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
